Neural activity persists during the maintenance of working memory (WM) representations and is thought to integrate perception and action over time and across brain areas through the coordination of multiple neural systems. Yet, there is a fundamental gap in understanding the neural mechanisms by which WM coordinates large-scale brain networks. This gap in knowledge is a critical problem because a host of psychiatric and neurologic symptoms stem from a primary WM dysfunction. The long-term goal of this work is to understand the mechanisms by which high-level cognition emerges through the temporal integration of sensory and motor functions across the cortex. The proposal's objective is to test new models of how the synchronization of neuronal oscillations may provide a neural mechanism for structuring recurrent interactions between different nodes in neural networks that support cognition. The central aim of the project is to test several critical predictions from recet theories of the role of neural oscillations and synchrony in high-level cognition using intracrania electroencephalography (iEEG) recordings from the prefrontal and posterior parietal cortices of human patients with pharmacologically intractable epilepsy. The rationale for the proposed research is that, as we better understand the mechanisms by which nodes in large-scale networks interact to give rise to high-level cognition, we will then be able to devise strategies fr understanding the basis, treatment, and prevention of mental disease. The objective will be to test, refine, and possibly refute, tenets of neural synchronization theories and will be accomplished by pursuing three specific aims: 1) Identify the frequencies at which neural oscillations persist during WM maintenance; 2) Test if WM maintenance enhances oscillatory frontal-parietal coupling; and 3) Determine how neural oscillations in different frequency bands interact. Strong preliminary data based on neural activity recorded from subdural electrodes on the surface of the frontal and parietal cortices of patients performing a memory guided saccade task demonstrate the feasibility of project aims in the applicant's hands. Under aim 1, gamma and alpha band oscillations were delay period (i.e., WM related) as well as spatially selective (i.e., contralateralized). Under aim 2, neural oscillations in frontal and parietal cortex synchronized during WM maintenance. Under aim 3, the phase of low frequency oscillations modulated the power of high frequency oscillations during WM maintenance. The approach is innovative because it capitalizes on an extremely rare population of patients with subdural electrodes over frontal and parietal cortex and relies on iEEG recording of neural signals that have the requisite sensitivity and temporal resolution to directly test recent theories of neural synchronization. The proposed research is significant because it is expected to test critical models of how neural oscillations structure computation and communication in the human brain thereby providing a thorough theoretical framework within which clinical researchers can develop strategies for the diagnosis and treatment of psychiatric and neurologic disorders.